基于颜色属性直方图的尺度目标跟踪算法研究

发布时间：2018-06-12 06:30

本文选题：目标跟踪 + 颜色属性　；参考：《电子与信息学报》2016年05期

【摘要】：利用目标颜色信息的跟踪算法,容易受到环境光照、尺度变化、相似背景等因素的干扰,导致跟踪任务失败。为了克服以上问题,该文提出一种基于颜色属性空间的鲁棒尺度目标跟踪算法。该算法首先将原始的RGB颜色空间映射到颜色属性(Color Names,CN)空间,减少目标颜色在跟踪过程中受环境变化影响。然后采用一种背景加权约束的颜色属性直方图,来抑制相似背景的干扰。最后,为了解决目标尺度变化带来的影响,先用梯度上升法粗略估计尺度,再用约束项精确求解尺度,并利用反向一致性检验,进一步提高尺度估计的准确性。该文选取了5段典型视频进行实验,并与相关算法进行比较。结果表明所提算法能够消除环境光照、阴影、相似背景和尺度变化等因素所带来的影响,在中心位置误差和跟踪成功率性能指标上,优于其它算法。
[Abstract]:The tracking algorithm based on the color information of the target is easy to be interfered by environmental illumination, scale change, similar background and so on, which leads to the failure of the tracking task. In order to overcome the above problems, this paper proposes a robust scale target tracking algorithm based on color attribute space. The algorithm firstly maps the original RGB color space to the color attribute Color NamesCNspace to reduce the influence of the environment on the target color in the tracking process. Then a color attribute histogram with background weighting constraints is used to suppress the interference of similar background. Finally, in order to solve the influence of the change of target scale, the scale is estimated roughly by gradient rising method, then the scale is solved accurately by the constraint term, and the accuracy of scale estimation is further improved by using the reverse consistency test. In this paper, five typical video segments are selected for experiments and compared with related algorithms. The results show that the proposed algorithm can eliminate the influence of environmental illumination, shadow, similar background and scale change, and is superior to other algorithms in the performance of center position error and tracking success rate.
【作者单位】：空军工程大学航空航天工程学院;空军工程大学空管领航学院;
【基金】：国家自然科学基金(61472442;61372167) 陕西省青年科技新星项目(2015KJXX-46)~~
【分类号】：TP391.41

【参考文献】